Getting Started with the MCP Server
BetterPost's MCP server lets you manage your entire content pipeline from any AI assistant. Add sources, generate content, update project settings, and more, all through natural conversation.
This guide walks you through connecting for the first time and running your first few commands.
What You Can Do
With the MCP server, your AI assistant can:
- Browse and manage stories discovered from your sources
- Generate content for any channel (newsletters, blog posts, social media)
- Add and remove sources to control what content you're monitoring
- Update project settings like audience, tone, and relevancy criteria
- Schedule recurring posts so content is generated automatically
- Save topics for future content generation
For a full list of tools, see the MCP Server reference.
Step 1: Connect Your AI Tool
Pick your AI tool and follow the setup instructions below. BetterPost uses OAuth, so there are no API keys to manage.
Claude Desktop
Add this to your config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"betterpost": {
"url": "https://betterpost.com/mcp"
}
}
}
Restart Claude Desktop after saving.
Claude Code
Run this in your terminal:
claude mcp add betterpost --transport http https://betterpost.com/mcp
VS Code (GitHub Copilot)
Add this to .vscode/mcp.json in your project or your user settings:
{
"servers": {
"betterpost": {
"type": "http",
"url": "https://betterpost.com/mcp"
}
}
}
Cursor
Add this to ~/.cursor/mcp.json:
{
"mcpServers": {
"betterpost": {
"url": "https://betterpost.com/mcp"
}
}
}
Windsurf
Add this to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"betterpost": {
"serverUrl": "https://betterpost.com/mcp"
}
}
}
Step 2: Sign In
The first time your AI assistant uses a BetterPost tool, a browser window will open asking you to sign in with your BetterPost account. Authorize the connection and you're set. The token is saved automatically so you won't need to sign in again.
Step 3: Explore Your Projects
Start by asking your AI assistant to list your projects:
"Show me my BetterPost projects"
This calls the list_projects tool and returns your project names and IDs. Pick a project to work with for the rest of this guide.
Step 4: Review Your Stories
Ask your assistant to show what content your sources have found:
"Show me the latest stories in my project"
Your assistant will call list_stories and return recent articles ranked by relevancy. Each story includes a title, source, publication date, and relevancy score. You can dig into any story for the full summary:
"Tell me more about that first story"
If a story isn't relevant, hide it so it won't be used for content generation:
"Hide that story, it's not relevant"
Step 5: Generate Your First Content
Now try generating content from your stories. You can generate for any channel:
"Write a LinkedIn post about the top story"
Or generate from a custom topic instead of a specific story:
"Write a newsletter about recent trends in AI developer tools"
The generate_content tool supports 10 channel types: newsletter, blog, twitter, linkedin, bluesky, facebook, email_outreach, tiktok, instagram, and youtube_shorts.
Each generation uses 1 credit, same as the web interface.
Step 6: Manage Your Sources
Add new content sources directly from your AI tool:
"Add https://techcrunch.com/feed as an RSS source to my project"
Or add a specific article as a story:
"Add this article to my project: https://example.com/interesting-article"
Check what sources you already have:
"List all sources in my project"
Common Workflows
Set Up a Weekly Newsletter
"Create a scheduled post for every Tuesday that generates a newsletter from trending topics"
This creates an automated schedule. BetterPost will generate the newsletter each week and you can review it before publishing.
Fine-Tune Your Project
Update your project settings to get better content:
"Update my project's target audience to 'engineering managers at mid-size startups'"
"Set additional instructions for LinkedIn posts: keep it under 200 words and end with a question"
"Update my relevancy criteria to focus on developer productivity and AI coding tools"
Save Topics for Later
When you come across a good topic but aren't ready to write about it yet:
"Save the topic 'how AI is changing code review workflows' for later"
Later, you can generate content from saved topics:
"Show my saved topics, then write a blog post about the first one"
Manage Your Content
Review what you've generated:
"Show me content I've generated recently"
"Delete that draft, I want to start over"
Tips
- Be specific about channels. Saying "write a LinkedIn post" is clearer than "write something about this."
- Use trending vs. recent topics. Trending topics are what's popular across your sources right now. Recent topics are chronologically newest.
- Combine stories. You can pass multiple story IDs to
generate_contentto create a roundup-style piece. - Check your usage. Ask "How many credits have I used?" to track your consumption.
Next Steps
- MCP Server Reference for the full list of tools and parameters
- Configuring Projects for detailed project settings
- Scheduled Posts for automating content generation
- Content Generator for content generation options in the web UI